Jump to content

  • Current Donation Goals

    • Monthly Donations
      Raised $440 of $700 target

So you're telling me there's a chance...

mrhonline

By mrhonline
in Homecourt

 Share

Recommended Posts

TheNorthCydeRises Grand Master

TheNorthCydeRises

Posted
Posted

The most interesting number to me:

15.1% chance of the Hawks making the finals

16.3% chance for Boston, 15.9% chance for Cleveland and a 15.9% chance for Orlando

Not much difference.

Yeah, that's what stands out to me. That, and the fact that the projections doesn't have any team in the league winning 60 or more games. I think that may change over time though.

Hawks record vs the top 10 teams in the league so far: 5 - 2

Link to comment
Share on other sites
Admin Newbie

Admin

Posted
Posted

For the most part, these playoff prediction things are a total waste of time.

Not when they're based on the numbers they aren't. This isn't someone's opinion on who will do what so it's completely unbiased and as he says in his article it was correct on all but 1 of the playoff teams at this point last year which is pretty damn good.

Link to comment
Share on other sites
  • Premium Member
Atlantaholic Grand Master

Atlantaholic

Posted
  • Premium Member
Posted

Not when they're based on the numbers they aren't. This isn't someone's opinion on who will do what so it's completely unbiased and as he says in his article it was correct on all but 1 of the playoff teams at this point last year which is pretty damn good.

The numbers are pretty meaningless because you are assuming that past performance will dictate future performance. Injuries, hot streaks, and a myriad of other variables are not taken into account. We have 100% chance of making the playoffs according to Hollinger, but that number means nothing. If Josh gets injured and we lose 4 straight than our playoff chances are not 100% anymore.

All that these playoff predictions tell you is: if every team continues to play exactly like they have been playing... this is the most likely scenario. Pretty freaking useless if you ask me.

Link to comment
Share on other sites
dlpin Newbie

dlpin

Posted
Posted

Not when they're based on the numbers they aren't. This isn't someone's opinion on who will do what so it's completely unbiased and as he says in his article it was correct on all but 1 of the playoff teams at this point last year which is pretty damn good.

I think at this point they are good for predicting playoff teams. But beyond that, it's a crapshoot. At this point last year Boston was by far the favorite to win 70 games, the east, and the championship.

Link to comment
Share on other sites
  • Premium Member
Atlantaholic Grand Master

Atlantaholic

Posted
  • Premium Member
Posted

Not when they're based on the numbers they aren't. This isn't someone's opinion on who will do what so it's completely unbiased and as he says in his article it was correct on all but 1 of the playoff teams at this point last year which is pretty damn good.

I could have predicted it just as good, and I don't need a formula to do it.

Link to comment
Share on other sites
buckeye242424 Newbie

buckeye242424

Posted
Posted

The numbers are pretty meaningless because you are assuming that past performance will dictate future performance. Injuries, hot streaks, and a myriad of other variables are not taken into account. We have 100% chance of making the playoffs according to Hollinger, but that number means nothing. If Josh gets injured and we lose 4 straight than our playoff chances are not 100% anymore.

All that these playoff predictions tell you is: if every team continues to play exactly like they have been playing... this is the most likely scenario. Pretty freaking useless if you ask me.

I agree with what you're writing, and it's true. Power rankings, future playoff seeds, who I should start on my fantasy football team this Sunday (yes, I'm in the playoffs) and next week...it's all moot. They all mean nothing. But, if the pattern we've seen through 21 games repeats, it's fun to think about the possibilities....and the numbers give it some merit.

Link to comment
Share on other sites
  • Premium Member
Atlantaholic Grand Master

Atlantaholic

Posted
  • Premium Member
Posted (edited)

I agree with what you're writing, and it's true. Power rankings, future playoff seeds, who I should start on my fantasy football team this Sunday (yes, I'm in the playoffs) and next week...it's all moot. They all mean nothing. But, if the pattern we've seen through 21 games repeats, it's fun to think about the possibilities....and the numbers give it some merit.

I'll give you that it can be fun, but the merit of the numbers really is not any better than the merit of any other "educated" opinion. In fact with opinions you at least have flexibility... You can account for the fact that the Spurs have been injured and usually surge in the second half, you can account for the fact that Utah had underperformed at the beginning of the year, you can account for the fact that Boston's age can catch up with them etc., either way it is all sheer speculation, and speculating that the trend from the first month of the season will continue throughout the entirety of the season is a hell of an assumption. It just bugs me a little that because Hollinger uses statistics that he pretends like it has any sort of extra validity... if statistics could predict the future I'd be making a fortune in the stock market and on sports bets.

I mean it's not like I need to run a lineal regresion to know that the Hawks are pretty good this year and have a chance at making the finals... I can just watch the games to determine that. Qualitative analysis, especially in something with the amount of variables like sports, is definitely just as valuable as quantitative analysis.

Edited by Atlantaholic
  • Like 1
Link to comment
Share on other sites
dlpin Newbie

dlpin

Posted
Posted

I mean it's not like I need to run a lineal regresion to know that the Hawks are pretty good this year and have a chance at making the finals... I can just watch the games to determine that. Qualitative analysis, especially in something with the amount of variables like sports, is definitely just as valuable as quantitative analysis.

I completely agree with this, and not only for basketball. I think in this obsession with statistics (and don't get me wrong, I do stats for a living) people tend to treat models like black boxes and don't check what is going in.

I mean, let's look at the model: there is nothing accounting for the fact that Orlando missed Lewis for a number of games because of a suspension that won't be there, there is nothing accounting for the fact that Doc is doing everything to preserve the health of his players, with most of the veterans playing near career lows in minutes, there is nothing accounting for the fact that certain teams have been unusually healthy (hawks, suns).

And this is one of the things I dislike about Hollinger (who I generally think if one of the better writers out there), the fact that he often uses statistics and statistical terms to write "sexier" conclusions than the data would allow him to.

Link to comment
Share on other sites
Guest

Guest

Posted
Posted

Not when they're based on the numbers they aren't. This isn't someone's opinion on who will do what so it's completely unbiased and as he says in his article it was correct on all but 1 of the playoff teams at this point last year which is pretty damn good.

Statistics like the ones Hollinger uses certainly are not unbiased. Statistics is still subject to biases, the use of numbers doesn't alleviate this. The big problem about Hollinger's playoff prediction stats? He doesn't post a formula, not does he explain the process well at all. Huge red flag to me.

Another red flag to me is its done by Hollinger. I probably sound like a broken record when Hollinger comes up in threads, but the guy isn't very good. His rogue use of statistics has spawned off many other scary writers like the dope Christopher Reina on RealGM. Its like they figured out Excel could create models for them and so they run to basketball-reference.com and fiddle around with models enough in order to create a stat that doesn't have much meaning except that it conforms to whatever their predetermined notion of player rankings should be. It wouldn't be very hard for me to go into Excel and create a ranking that ends up with Mike Bibby as the #1 ranked player and then for me to slap on Fanatical Value to the statistic I used. Then because most people shy away from math, they just take the stat and either agree/disagree based on how cool of an explanation there is behind it.

But mrH, what was all that one in a million talk?

Link to comment
Share on other sites
DJlaysitup Newbie

DJlaysitup

Posted
Posted (edited)

"Figures don't Lie....but Liars Figure"

It's akin to the BCS IMO. They say "the computer rankings"...as if the computer is all knowing and has a mind of it's own. We all know what a computer is without human programmed software - an expensive rock.

Edited by DJlaysitup
Link to comment
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
 Share
Go to topic listing

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...